Infrastructure: Real Assets and Real Returns

• Author: Harry Liem,Ron Bird,Susan Thorp
• Published date: 01 September 2014
• DOI: http://doi.org/10.1111/j.1468-036X.2012.00650.x
• Date: 01 September 2014

European Financial Management, Vol. 20, No. 4, 2014, 802–824

doi: 10.1111/j.1468-036X.2012.00650.x

Infrastructure: Real Assets and Real

Returns

Ron Bird

Paul Woolley Centre, University of Technology, Sydney, PO Box 123 Broadway NSW 2007, Australia

E-mail: ron.bird@uts.edu.au

Harry Liem

University of Technology, Sydney, PO Box 123 Broadway NSW 2007, Australia

E-mail: harry.liem@student.uts.edu.au

Susan Thorp

University of Technology, Sydney, PO Box 123 Broadway NSW 2007, Australia

E-mail: susan.thorp@uts.edu.au

Abstract

Little empirical work has been done on infrastructure as an asset class despite

increased allocations by institutional investors. We build a robust factor model of

infrastructurereturns using US and Australian infrastructureand utility data to test

manager claims that infrastructure investments offer benefits via a combination of

monopolistic and defensive assets. We find evidence of excess returns and inflation

hedging, but not of defensive characteristics. We compare option-based models

designed to replicate infrastructure asset returns, and identify the regulatory risk

premium. A combination of inflation linked bonds and covered call strategies

results in improved defensive and inflation hedging characteristics.

Keywords: inflation hedging,regulatory risk premium,infrastructure assets.

JEL classification: G12, G14

The authors wish to acknowledge the generous support of the Paul Woolley Centre at

the University of Technology, Sydney (UTS). In addition, Thorp acknowledges Australian

Research Council (ARC) DP 0877219. The Chair of Finance and Superannuation at UTS

receives support from the Sydney Financial Forum (Colonial First State Global Asset

Management), the NSW Government, the Association of Superannuation Funds of Australia

(ASFA), the Industry Superannuation Network (ISN), and the Paul Woolley Centre, UTS.

The authors also wish to thank John Doukas, two anonymous referees, and attendees at the

2011 Paul Woolley Conference in Sydney for their valuable comments and suggestions.

C

2012 Blackwell Publishing Ltd

© 2012 John Wiley & Sons Ltd

Infrastructure 803

© 2012 John Wiley & Sons Ltd

1. Introduction

Pension funds around the world have been investing in infrastructure with the aim of

providing better long duration, real (inflation adjusted) retirement benefits to members.

While infrastructure has been seen as a suitable institutional investment in Australia,

Canada and the UK since the 1990s, it has only recently increased in popularity in the

USA and, unlike hedge funds or private equity, has not been the focus of much academic

research, partly because of limited data (Dechant et al., 2010; Martin, 2010; Newell and

Peng, 2008).

In this paper we define infrastructure as encompassing the utility sector (power

generation and distribution) as well as the ‘pure’ infrastructure sector (toll roads, com-

munications and airports). Institutional investors most commonly delegate infrastructure

investment to managers but may also undertake direct investment (Inderst, 2009).1We

consider investment in both listed and unlisted Australian infrastructure assets, either

directly or through managed funds. Outside of Australia, where less data is available,we

focus on the listed infrastructure markets.

Initially researchers viewed infrastructure as having characteristics akin to real estate:

a long life cycle, heterogeneity, illiquidity and inflation hedging ability (Dechant et al .,

2010; Newell and Peng, 2007, 2008). However, a defining feature of the sector is the

prevalence of ‘natural monopolies’ in the provision of essential services (such as energy

or water) and of economies of scale created bylarge distribution networks (Dimasi, 2010).

Researchers now concur that infrastructure is a distinct asset class having oligopolistic

or monopolistic characteristics, because of restrictions on ownership and on the uses to

which infrastructure can be put. Natural monopolies and industries enjoying economies

of scale are often regulated by government restrictions on prices, returns, output levels

or barriers to entry. Consequently, infrastructure investments require compensation for

regulatory risk. In many countries, infrastructure consists of mature utilities, increasingly

facing deregulation and privatisation (Inderst, 2009). As a result, the ability to deliver

superior performance may diminish as the industry matures (deregulates).

We makethe following three contributions to literature. First, we suggest style factors

common to both USA and Australian infrastructure investments. We extend our analysis

by applying an asymmetric GARCH model to take account of observed volatility

clustering in excess returns and we test both the defensive and the inflation hedging

ability of the asset class. We find evidence of excess returns and inflation hedging

features in infrastructure investments in the USA and Australia. While we do find some

desirable characteristics, we caution against sample size and time period biases given

the limitations of our data set.

Second, we test the defensive ability of infrastructure investments during equity down

markets. We find no evidence of defensive characteristics. Despite claims that it is a

defensive sector, Australian listed infrastructure funds performed very poorly during the

Global Financial Crisis.2

1Infrastructure investment methods are described in detail in the appendix.

2Notably, large infrastructure f irms such as Macquarie and Babcock & Brown werecriticised

for ‘asset gathering’ using leverage, thereby generating transaction, advisory, base and

performance fees for the parent company. In March 2009, Babcock and Brown was placed

into voluntary administration due to its over-reliance on short-term debt while Macquarie

restructured and survived as Intoll Group.